Page 1 of 1

Basque is IE!?

Posted: Tue Jun 23, 2015 2:33 am
by garysk
Just saw on Amazon a book entitled "A First Etymological Dictionary of Basque as an Indo-European Language: Basque Native, Basic Lexicon" by Gianfranco Forni, claiming to have "discovered" the sound changes that lead from IE to Basque. Has anyone here seen or read or heard of this? Is it quackery, or is it real? Not terribly expensive ($19.00 at Amazon),, but don't want to waste money on a crackpot. I searched this forum for "Forni" and got no results, so I assume no one has posted about it.

Re: Basque is IE!?

Posted: Tue Jun 23, 2015 2:36 am
by GreenBowTie
it's almost certainly the work of a quack, but i'd be curious to read it too!

Re: Basque is IE!?

Posted: Tue Jun 23, 2015 4:33 am
by Zju
I've stumbled upon this guy some time ago and maybe even posted about him in the linguistic quackery thread. In his academia.edu profile he says that his long term mission is to find the connection between all languages or something. He says he is not prejudiced with the outcome, but it's obvious he is trying to lump together as many languages as possible in as few macro-families as possible. So yeah, borderline crackpot at best.
Unlike most other cracpots, however, his methodology is solid at least at first sight. Though I'm not sure how much this helps if you have predetermined goal in mind. Also unlike other crackpots, he actually leads dialogue with his critics.
What makes me doubt the most in his reconstructions of Basque as IE is that he posits that more than half of word initial consonants eventually dropped, which makes finding cognates all the easier.
If you want to read his works, which are interesting at times, you can find them at academia.edu. I imagine the book is mostly all of his articles put together plus maybe some more research not online.

Re: Basque is IE!?

Posted: Tue Jun 23, 2015 7:29 am
by Salmoneus
There's a simple test of whether a book is written by a crackpot:
- does it contain the word 'Basque' in the title?
- if so, do they mean 'Basque' the language, rather than the people?
- if so, is it a grammar of Basque?
- if not, there is a 99% chance it is written by a crackpot

Re: Basque is IE!?

Posted: Tue Jun 23, 2015 1:50 pm
by vokzhen
As Zompist (I think?) has pointed out before, you can make sound changes for a lot of languages, but that doesn't mean they're right. What was the example he used - Quechua and Mandarin? Wakashan and English? Something crazy like that. You *can* make lists of sound laws that give seemingly-good correspondences, but there's a reason morphological correspondences are held in higher regard. Things like Indo-European and Uralic both having merged case-and-number suffixes and sound correspondences between them, across Na-Dene there's a set of immediately preverbal transitivity-modifying "classifiers" and very similar order of verb prefixes (which are two of the arguments for Dene-Yenesian as well), and the first example in the Sapir's paper proposing uniting Algonquian with Yurok and Wiyot is the shared set m- n- k- w- for marking unpossessed, 1S, 2S, and 3S body part possession.

Re: Basque is IE!?

Posted: Tue Jun 23, 2015 2:20 pm
by Pabappa
A few years ago someone posted a link here to a study claiming that Burushaski was actually IE, despite the study basically having no evidence. However I'd say Basque being IE is even more pathetic since we can trace Basque back at least a few thousand years and it actually gets less IE-looking as it goes, while narrowing down the supposed time of divergence to just a few thousand years. Essentially, nothing matches whatsoever, not the number system, nor the pronouns, nor the grammar. It just doeesnt work. I think Basque is actually *related* to IE, but at a time scale so deep that we'll never be able to actually make the connection even if somehow someone discovers an ancient script written by a lost civilization in 8438 BC desrtbining all languages of Europe & Asia in extreme detail.

Re: Basque is IE!?

Posted: Tue Jun 23, 2015 2:25 pm
by garysk
Yeah, thanks guys. This is exactly what I thought. But I ordered the book anyway, out of curiosity. I also ordered (guess I'm on a crackpot binge) "Discoveries in Hebrew, Gaelic, Gothic, Anglo-Saxon, Saxon, Latin, Basque and Other Caucasic Languages Showing Fundamental Kinship of the Aryan Tongues ... With the Semitic Tongues", by Allison Emery Drake. The hodge-podge title, and the bargain price, tell me this one is also cracked. But it might be amusing as well. While it seems reasonable to assume that language started somewhere, and that the ones who first had it imparted it to descendants, that was sooo long ago, the relationships are lost forever, undiscoverable, barring time travel (not holding my breath). This written just as Publipis' post was coming in.

Re: Basque is IE!?

Posted: Tue Jun 23, 2015 6:25 pm
by Richard W
Zju wrote:What makes me doubt the most in his reconstructions of Basque as IE is that he posits that more than half of word initial consonants eventually dropped, which makes finding cognates all the easier.

Forni actually got published in a serious academic journal so that his idea and, more importantly, method could be discussed.

The one thing he could actually be nailed on was unreasonable modes of word formation in the hypothesised PIE-like precursor Basque. The rest was an embarrassing revelation that we don't have explicit criteria for dismissing implausible sets of sound changes.

The criticism that Zju makes was actually the unfairest of them all. The idea seems to have been originally proposed by André Martinet, and is based on internal evidence. To be precise, the idea is that initial /p/, /t/ and /k/ were lost, the trajectory being something like stop > aspirate > fricative > /h/ > zero.

Re: Basque is IE!?

Posted: Sun Jun 28, 2015 6:48 am
by Zju
Richard W wrote:
Zju wrote:What makes me doubt the most in his reconstructions of Basque as IE is that he posits that more than half of word initial consonants eventually dropped, which makes finding cognates all the easier.

Forni actually got published in a serious academic journal so that his idea and, more importantly, method could be discussed.

The one thing he could actually be nailed on was unreasonable modes of word formation in the hypothesised PIE-like precursor Basque. The rest was an embarrassing revelation that we don't have explicit criteria for dismissing implausible sets of sound changes.

The criticism that Zju makes was actually the unfairest of them all. The idea seems to have been originally proposed by André Martinet, and is based on internal evidence. To be precise, the idea is that initial /p/, /t/ and /k/ were lost, the trajectory being something like stop > aspirate > fricative > /h/ > zero.


I'm not concerned by the loss of initial /p/ /t/ /k/ per se - rather by the large number of sound changes that lead to simplification and in turn, make finding false positives easier.

Consider this: You have the hypothetical Proto-Basque word *VsV - it could point to any of PIE *VsV, *pVsV, *tVsV, *kVsV, *ḱVsV, *kʷVsV, *mVsV - that's 7 roots that could *VsV possibly be reflex of, you only need to have similar semantics. But wait, intervocalic *s could be reflex of either *s or *st - that makes 14 candidate roots. And then it could be a reflex of a root with initial consonant cluster: *CsV > *CVsV > *VsV - that makes even more potential PIE sources for *VsV. You do the math. That's even before we've taken the loss of laryngeals into consideration.

Using the sound laws he has posited, each Proto-Basque word has many potential PIE sources, many more than ten.

As Zompist (I think?) has pointed out before, you can make sound changes for a lot of languages, but that doesn't mean they're right. What was the example he used - Quechua and Mandarin? Wakashan and English?

It doesn't happen to be this and this pages, does it? They're only about lookalikes, not about sound changes deriving set of words from language X to Y. I'd be very interested to read about the latter, if someone ever manages to demonstrate it's possible.

Re: Basque is IE!?

Posted: Sun Jun 28, 2015 6:54 pm
by Richard W
Zju wrote:Consider this: You have the hypothetical Proto-Basque word *VsV - it could point to any of PIE *VsV, *pVsV, *tVsV, *kVsV, *ḱVsV, *kʷVsV, *mVsV - that's 7 roots that could *VsV possibly be reflex of, you only need to have similar semantics. But wait, intervocalic *s could be reflex of either *s or *st - that makes 14 candidate roots. And then it could be a reflex of a root with initial consonant cluster: *CsV > *CVsV > *VsV - that makes even more potential PIE sources for *VsV. You do the math. That's even before we've taken the loss of laryngeals into consideration.

Using the sound laws he has posited, each Proto-Basque word has many potential PIE sources, many more than ten.

The amazing thing is that more people haven't availed themselves of these changes in trying to relate Basque to other groups. Some languages do a very good job of preventing a convincing demonstration that they are related to anything else.